[Debichem-devel] Bug#1042048: python-ase: FTBFS: E: Build killed with signal TERM after 150 minutes of inactivity

[Santiago Vila]

tags 1042048 - unreproducible forky sid trixie
tags 1042048 + bookworm
thanks

Hi.

The original build log from Lucas is now removed. Lucas: I wonder if
it would be possible for you to keep those build logs for more time,
or maybe keep them until the issue is solved.

I am able to reproduce this build failure 100% of the time on AWS machines
which incidentally have a single CPU, but I'm unable to determine if
there is a relation at all between failure and number of available
CPUs (correlation != causation).

If that was indeed the root cause, it would be possible
to "import os" and then skipif(os.cpu_count()==1).

If we are unable to determine the root cause, I would consider the test
as flaky and would disable it (in bookworm) without any "if".

As always, I will be more than happy to provide a VM where this
happens all the time to anybody willing to debug it.

Thanks.